def test_undefined_dataset():
with pytest.raises(AttributeError) as err:
data("blahblahblah")
assert str(err.value) == "No dataset named 'blahblahblah'"
with pytest.raises(AttributeError) as err:
local_data("blahblahblah")
assert str(err.value) == "No dataset named 'blahblahblah'"
def get_cars_asp():
ch = Chart(data('cars'))
ch = ch.field('Horsepower', 'Displacement')
qu = ch._get_full_query()
for l in qu:
print(l)
def test_load_local_dataset(name):
loader = getattr(data, name.replace("-", "_"))
local_loader = getattr(local_data, name.replace("-", "_"))
df1 = data(name)
df2 = loader() # equivalent to data.dataset_name()
df3 = local_data(name)
df4 = local_loader() # equivalent to local_data.dataset_name()
assert_frame_equal(df1, df2)
assert_frame_equal(df1, df3)
assert_frame_equal(df1, df4)
raw1 = loader.raw()
raw2 = local_loader.raw()
raw3 = data(name, return_raw=True)
raw4 = local_data(name, return_raw=True)
assert raw1 == raw2 == raw3 == raw4
assert type(raw1) is type(raw2) is type(raw3) is type(raw4) is bytes
def index(self):
"""Render the plots"""
plots = []
df = data('iris')
url = plot_iris_scatter(df)
plots.append({
'title': 'Scatterplot',
'url': url
})
df = data('iris')
url = plot_iris_histogram(df)
plots.append({
'title': 'Histogram',
'url': url
})
return {
'plots': plots
}
def api(self):
"""Data for the plots"""
dataset = self.request.params.get('dataset')
if not dataset:
raise ValueError('"dataset" value must be given')
if dataset not in data.list_datasets():
raise ValueError('Dataset "{}" not found'.format(dataset))
df = data(dataset)
return df.to_dict(orient='list')
def index(self):
"""Render the plots"""
plots = []
# Build plot
df = data('iris')
# source = ColumnDataSource(df)
source = AjaxDataSource(data_url='/api/bokeh?dataset=iris')
species = df['species'].unique()
fill_color = factor_cmap('species', palette=Category10[len(species)], factors=species)
hover = HoverTool(tooltips=[
('index', '$index'),
('petalLength', '@petalLength{0.0}'),
('sepalLength', '@sepalLength{0.0}'),
('species', '@species'),
])
p = figure(title='Iris Morphology', tools=[
hover, 'pan', 'wheel_zoom', 'reset'], active_scroll='wheel_zoom')
p.xaxis.axis_label = 'petalLength'
p.yaxis.axis_label = 'sepalLength'
p.circle(x='petalLength', y='sepalLength', color=fill_color,
fill_alpha=0.5, size=10, source=source)
# Save plot
script, div = components(p)
basename = 'iris_scatter'
filename = '{}/{}.js'.format(PLOTS_DIR, basename)
url = to_url(filename)
file = open(filename, 'w')
file.write(etree.fromstring(script).text)
file.close()
plots.append({
'title': 'Scatterplot',
'div': div,
'script_url': url
})
return {
'plots': plots
}
def index(self):
"""Render the plots"""
plots = []
# np.random.seed(sum(map(ord, 'calmap')))
# all_days = pd.date_range('1/15/2014', periods=1200, freq='D')
# days = np.sort(np.random.choice(all_days, 50))
# # Year heatmap
# events = pd.Series(np.random.randn(len(days)), index=days)
# path, url = plot_year_heatmap(events)
# plots.append({
# 'title': 'Year heatmap',
# 'url': url
# })
# # Calendar heatmap
# events = pd.Series(np.random.randn(len(days)), index=days)
# path, url = plot_calendar_heatmap(events)
# plots.append({
# 'title': 'Calendar heatmap',
# 'url': url
# })
# Boxplot
df = data('flights-2k')
first_5 = df['origin'].unique()[:5]
df = df[df['origin'].isin(first_5)]
df['date'] = pd.to_datetime(df['date'])
df['weekday_abbr'] = df['date'].dt.strftime('%a')
df['weekday_rank'] = df['date'].dt.weekday
df.sort_values('weekday_rank', inplace=True)
with sns.axes_style('whitegrid'):
ax = sns.boxplot(x='weekday_abbr',
y='delay',
hue='origin',
data=df,
dodge=True,
linewidth=1,
flierprops={
'markersize': 5,
'marker': '.'
})
# ax = sns.swarmplot(x='weekday_abbr', y='delay', hue='origin', data=df,
# dodge=True, color='.3', size=3)
sns.despine(offset=10, trim=True)
svg = save_figure(ax.get_figure(), SVG, 900, 450)
# df = data('iris')
# with sns.axes_style('whitegrid'):
# ax = sns.scatterplot(x='petalLength', y='sepalLength', hue='species',
# palette='Spectral', data=df)
# svg = save_figure(ax.get_figure(), SVG, 400, 300)
plots.append({'title': 'Scatterplot', 'svg': svg})
# Scatterplot
# df = data('iris')
# path, url = plot_iris_scatter(df)
# plots.append({
# 'title': 'Scatterplot',
# 'url': url
# })
return {'plots': plots}
def get_cars_asp():
ch = Chart(data('cars'))
ch = ch.field('Horsepower', 'Displacement')
qu = ch._get_full_query()
for l in qu:
print(l)
if __name__ == "__main__":
# get_cars_asp()
# exit(1)
gapminder = data('gapminder')
chart = Chart(gapminder)
schema = chart._get_full_query()
print('\n\n---- HERE IS THE DEFAULT VIEW AND SCHEMA ----')
for l in schema:
print(l)
gapminder1 = Chart(gapminder).field('pop',
scale='zero').field('life_expect',
scale='zero')
# get the default view and schema asp.
print('\n\n---- HERE IS THE DEFAULT VIEW, SCHEMA PLUS THE QUERY ----')
query = gapminder1._get_full_query()
for l in query:
print(l)
def test_date_types(name, col):
assert data(name)[col].dtype == "datetime64[ns]"
#!/usr/bin/env python
# coding: utf-8
# In[1]:
from dziban.mkiv import Chart
from vega_datasets import data
from copy import deepcopy
import json
from tqdm import tqdm
import pandas as pd
# In[2]:
movies = data('movies')
movies.head(1)
# In[3]:
base = Chart(movies)
# In[4]:
q1 = 'IMDB_Rating'
q2 = 'Rotten_Tomatoes_Rating'
q3 = 'Worldwide_Gross'
n1 = 'Major_Genre'
n2 = 'MPAA_Rating'
n3 = 'Creative_Type'
fields = {'q': [q1, q2, q3], 'n': [n1, n2, n3]}
"""
Stacked Bar Chart
-----------------
This example shows how to make a stacked bar chart of the weather type in Seattle from 2012 to 2015 by month.
"""
import altair as alt
from vega_datasets import data
weather = data('seattle-weather')
chart = alt.Chart(weather).mark_bar().encode(
color=alt.Color(
'weather:N',
legend=alt.Legend(title='Weather type'),
scale=alt.Scale(
domain=['sun', 'fog', 'drizzle', 'rain', 'snow'],
range=['#e7ba42', '#c7c7c7', '#aec7e8', '#1f77b4', '#9467bd']),
),
x=alt.X(
'date:N',
axis=alt.Axis(title='Month of the Year'),
timeUnit='month',
),
y='count(*):Q',
)
def load_dataset(name):
"""Load a dataset by name as a pandas.DataFrame."""
import vega_datasets
return vega_datasets.data(name)
def animate(i):
print(i)
ax.clear()
return plot()
anim = FuncAnimation(fig, animate, frames=range(10), interval=30)
plt.show()
###
from sklearn.linear_model import LinearRegression
from pydataset import data
import textwrap as tw
fig, ax = plt.subplots()
ax.set(xlim=(0, 45), ylim=(0, 8))
tips = data('tips').sample(8, random_state=456)
lm = LinearRegression().fit(tips[['total_bill']], tips.tip)
m = lm.coef_[0]
b = lm.intercept_
x = tips.total_bill
y = tips.tip
yhat = m * x + b
def animate(i):
if i == 0:
ax.clear()
ax.plot(x, y, ls='', marker='o', label='actual ($y$)', color='black')
ax.set(xlim=(0, 45), ylim=(0, 8), title='Our orignal X and Y data')
ax.legend(loc='upper left')
elif i == 1:
ax.set(title='Add our line to make predictions, y = {:.2f}x + {:.2f}'.format(m, b))
ax.plot(x, yhat, color='grey', label='')
def load_dataset(name):
"""Load a dataset by name as a pandas.DataFrame."""
warnings.warn(MSG, DeprecationWarning)
import vega_datasets
return vega_datasets.data(name)